On October 20, 2025, large parts of the internet slowed to a crawl—or went dark entirely—for several hours. The common thread was Amazon Web Services (AWS), the cloud platform that powers thousands of apps and websites. When AWS stumbled, popular services from messaging to banking felt the impact worldwide. (Reuters)

A quick recap of the day

• When: Early Monday, Oct 20, 2025 (globally; local times varied).
• What: A major AWS outage caused high error rates and latency across core services.
• Who was affected: Apps like Snapchat, Signal, Fortnite, Roblox, Venmo, Coinbase, and even Amazon’s own Alexa/Prime Video/Ring—plus some banks and government portals. (Reuters)

The likely cause (in plain English)

AWS reported problems centered in its US-East-1 region (Northern Virginia), a huge hub many companies rely on. Reporting pointed to issues in AWS’s internal traffic management—think of the system that routes digital “cars” on the internet’s highways—and, in some coverage, to DNS components that translate web names into numeric addresses. The upshot: requests couldn’t reach the right place reliably, so apps failed to load. While details vary across outlets, no evidence of a cyberattack was found. (The Guardian)

Why one region can ripple worldwide

Many companies design their apps to lean heavily on a single AWS region for speed and simplicity. When that region hiccups, the effect can be global because:

1. traffic gets rerouted or fails,
2. shared services (like authentication or content delivery) stumble, and
3. dependent apps time out—even if your phone or local ISP is fine. (The Guardian)

Who felt it the most?

• Social & chat: Snapchat and Signal saw spikes in outage reports.
• Gaming & media: Fortnite, Roblox, and streaming features lagged or went offline.
• Money apps: Coinbase, Robinhood, Venmo, and some UK banks (e.g., Lloyds/Halifax) had disruptions.
• Everyday devices: Ring doorbells and Alexa voice commands misbehaved for many users. (Reuters)

How AWS and apps recovered

Engineers began mitigation within hours, gradually restoring services. As fixes rolled out in the affected region, error rates dropped and apps came back—some faster than others depending on how they’re built. Outages of this nature are disruptive but not unprecedented in large cloud systems. (AP News)

What you can do next time

You can’t prevent a cloud provider outage, but you can reduce the pain:

• Have alternatives: If one messaging app dies, switch to SMS or another platform.
• Download offline essentials: Maps, tickets, and 2FA backup codes.
• Know service status pages: AWS Status (for tech teams) and the status pages of apps you rely on.
• For teams/IT:
  • Spread workloads across multiple regions/providers.
  • Test failover regularly.
  • Keep read-only modes so customers can still see content when writes fail.
  • Cache critical data (e.g., user profiles) to survive brief upstream outages.

A simple example

Imagine an online store that keeps its product catalog cached and uses a backup payment gateway. If its primary cloud region fails, shoppers can still browse (cached) and pay (fallback gateway), even if non-critical features like recommendations are temporarily disabled. That’s graceful degradation in action.

The bigger lesson

October 20 underscored how much of the web depends on a few cloud giants. Centralization brings speed and cost benefits—but also creates single points of failure. Regulators and experts are increasingly calling for diversification and resilience standards so everyday services (including public-sector sites) don’t all blink out at once. (The Guardian)

Bottom line: This wasn’t your Wi-Fi. It was a large cloud hiccup with global consequences—and a fresh reminder to design for resilience. (Reuters)